Marginal predation methodologies and the importance of predator preferences

Correspondence: T. Stankowich, Psychology Department, Young Hall, One Shields Avenue, University of California, Davis, CA 95616, U.S.A. (email: [email protected]). Biologists have studied group dynamics and attempted to define the benefits of gregariousness across diverse taxa for several decades. The relative costs and benefits of group living (reviewed in Krause & Ruxton 2002) may differ between group members so that each is under selection to maximize its individual fitness in different ways (Rubenstein 1978). For example, individuals on the edge of a group may have higher feeding rates but suffer higher predation risk, whereas those on the inside enjoy lower rates of predation but suffer reduced feeding rates (e.g. colonial spiders: Rayor & Uetz 1990). Many years ago ornithologists studying predation rates in colonialnesting birds reported that predation rates on eggs or fledglings were higher for solitary nests outside of the colony than for those inside the colony, and positions on the periphery of the colony were more susceptible to predation than central positions (e.g. Taylor 1962; Kruuk 1964; Patterson 1965). This phenomenon of differential predation risk within a group (i.e. a greater risk of predation on the periphery relative to the centre) is known as ‘marginal predation’. Similar early studies found the same phenomenon of differential risk in mobile animals (e.g. mammals: Galton 1871; birds: Wynne-Edwards 1962). Following these early studies, several mathematical and geometrical models of the benefits of group formation emerged, the most famous of which is Hamilton’s (1971) theory of the ‘selfish herd’, in which aggregations of animals form purely by selfish behaviour of individuals. Hamilton’s (1971) predictions were echoed and extended by Vine (1971) and since have been refined and formalized by computer models to show how within-group movements of individuals during attacks may minimize their individual predation risk (e.g. Morton et al. 1994; Gueron et al. 1996; Barta et al. 1997; Beecham & Farnsworth 1999; Viscido et al. 2001). While Hamilton’s (1971) ideas have been the inspiration behind most of the theoretical issues at hand, the problem with analysing the empirical validity of these models is that we do not have standardized methodologies for measuring marginal predation and benefits of aggregation.